TPR FL1 B/B Q#5 "SPOILER"

[betterfuture]

I have the slightest clue on what I am supposed to figure out or what this answer tells me. Someone willing to try?

---------------------------------------------------------------------------------------------------------------------

---

Members don't see this ad.

 

[aldol16]

You're only looking at the RR and not at the confidence interval. The confidence interval is important because it's basically the "error bar." So if I told you that something had a mean of 0.75 with an error bar spanning from 0.5 to 1.24, you could not say with confidence that the true mean is not 1.0. The only case where you could say that the true mean is not 1.0 with >95% confidence is with urea nitrogen.

 

[betterfuture]

Hmm. I am not sure I quite get it yet. Could you explain further? Never was good at statistics

 

[aldol16]

Are you familiar with error bars? That's probably the easiest way to look at this. Say you're doing an experiment on obesity. You feed rats X, Y, or Z and compare their weights to controls after 15 weeks. You plot a bar graph that shows the fold change in weight for each diet relative to controls. So control is 1.0, X is 1.4, Y is 1.2, and Z is 2.0 (made these numbers up arbitrarily). To make it clear, this means that if the control rat weighs 10 g, then the rat eating diet X weighs 1.4 grams and so on. Okay, so your bar graph will show four bars with the following heights: 1.0, 1.4, 1.2, and 2.0. Okay, can you draw any conclusion from this?

NO! You can't draw a conclusion because you don't know the error in your measurement. So I'll give you that too - this is called a confidence interval. And I'm giving you the 95% confidence interval, which is standard. This means that if you were to repeat the measurement 100 times, the true value will lie inside that interval 95 times. Okay, so X becomes 0.9 - 1.6, Y is 0.8 - 1.9, and Z is 1.5 - 2.5. So in your mind, insert those error bars into your bar graph. Here's the question: did any diet significantly alter the rat's weight?

 

[betterfuture]

So in all cases, X, Y, and Z, the rats did lose weight but is it significant? Is that what your asking? And the true value is the average weight loss? So basically, in all cases, the true value should lie inside each interval, right? And you can say that with 95% confidence. The 5% represents the error region? Not sure.

 

[aldol16]

What I'm saying is that in all cases except Z, the rats did not gain weight. It's critical that you understand this. They did not gain weight because 1.4 and 1.2 are not significantly different from 1.0 given their respective confidence intervals.

Think of it this way. If I showed you two bar graphs with error bars that overlapped, would you say there's a significant difference? By now, you should be able to answer no. Again, this is crucial for the B/BC section. You must be able to interpret statistical results both graphically and numerically. Here, I'm saying that the error bars for X range from 0.9 to 1.6. In other words, these error bars would overlap with control (1.0) and thus you can't say that there's a significant difference between the two. Ergo, X did not gain weight.

 

[betterfuture]

Thank you aldol16! I'm going to have to go over some statistics to get the hang of it!